arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2211.15876cs.CV
收藏

特定于实例的图像目标导航:训练具身智能体查找对象导航

Instance-Specific Image Goal Navigation: Training Embodied Agents to Find Object Instances

Oregon State University(俄勒冈州立大学);Meta AI;UC Berkeley(加州大学伯克利分校);Georgia Tech(佐治亚理工学院)

Jacob Krantz, Stefan Lee, Jitendra Malik, Dhruv Batra, Devendra Singh Chaplot

摘要:

我们考虑的问题,具体的视觉导航给定一个图像的目标(ImageNav),代理初始化在一个不熟悉的环境中,并负责导航到一个位置的图像“描述”。与相关的导航任务不同,ImageNav没有标准化的任务定义,这使得在方法之间进行比较变得困难。此外,现有的公式具有两个有问题的性质:(1)图像目标是从随机位置采样的,这可能导致模糊性(例如,看着墙壁),以及(2)图像目标与照相机规格和代理的实施例相匹配;当考虑用户驱动的下游应用时,这种刚性是有限的。我们提出了特定于实例的ImageNav任务(InstanceImageNav)来解决这些限制。具体地说,目标图像被“聚焦”在场景中的某个特定对象实例上,并且使用独立于代理的相机参数来拍摄。我们在Habitat Simulator中实例化了InstanceImageNav,使用了来自Matterport 3D数据集(HM 3D)的场景,并发布了一个标准化的基准来衡量社区的进展。

英文摘要:

We consider the problem of embodied visual navigation given an image-goal (ImageNav) where an agent is initialized in an unfamiliar environment and tasked with navigating to a location 'described' by an image. Unlike related navigation tasks, ImageNav does not have a standardized task definition which makes comparison across methods difficult. Further, existing formulations have two problematic properties; (1) image-goals are sampled from random locations which can lead to ambiguity (e.g., looking at walls), and (2) image-goals match the camera specification and embodiment of the agent; this rigidity is limiting when considering user-driven downstream applications. We present the Instance-specific ImageNav task (InstanceImageNav) to address these limitations. Specifically, the goal image is 'focused' on some particular object instance in the scene and is taken with camera parameters independent of the agent. We instantiate InstanceImageNav in the Habitat Simulator using scenes from the Habitat-Matterport3D dataset (HM3D) and release a standardized benchmark to measure community progress.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑